This proposal, "Pilot trial of metformin in amnestic MCI", is a resubmission of R01 AG026413-01A1 in response to PAR-07-142 (Alzheimer's disease pilot clinical trials). Hyperinsulinemia has surfaced as a potential risk factor for Alzheimer's disease (AD) based on epidemiological, clinical, and laboratory studies. Half of the population 60 years and older may have hyperinsulinemia, and the prevalence is increasing with the epidemic of overweight and obesity. This has prompted clinical trials of an insulin lowering agent, rosiglitazone, in the treatment of AD. However, recent concerns on the safety of rosiglitazone make its utility uncertain. We propose to test another medication that lowers insulin levels, metformin. Metformin is being used safely and effectively for the treatment of diabetes, and to reduce insulin levels and prevent diabetes in persons at risk. Our hypothesis is that metformin can prevent cognitive decline by reducing insulin levels in overweight and obese persons (without diabetes) with amnestic mild cognitive impairment (AMCI), who are at a high risk of progressing to Alzheimer's disease. We propose to conduct a phase II placebo-controlled randomized trial of metformin 1000 mg twice a day in 60 persons with AMCI lasting 12 months. The primary goals of this trial are to conduct an assessment of efficacy of metformin in AMCI and to collect preliminary data for establishing clinical and neuroimaging measures of efficacy for use in a phase III trial, consistent with the PAR. The primary aim of this study is to determine whether metformin reduces the risk of cognitive decline, measured with the modified ADAS COG, in persons with AMCI compared to placebo at 12 months based on an intention to treat analysis. Our secondary aim is to determine whether metformin prevents the decrease in brain metabolism, characteristic of the transition from AMCI To AD, in the posterior cingulate cortices between baseline and 12 months measured with brain [18]F-labeled 2-deoxy-2-fluoro-D-glucose (FDG) positron emission tomography (PET) with magnetic resonance imaging (MRI) co-registration compared to placebo in 40 persons in an imaging substudy. The results of this study will inform the decision to proceed to a phase III trial of metformin in AMCI, and provide information of effect estimates in the outcomes for calculation of sample size for a phase III trial. Given the dual epidemics of hyperinsulinemia and AD it is important to determine whether interventions that reduce insulin levels can prevent cognitive decline and AD, and metformin is a safe and effective candidate. This is an example of translational research and is consistent with the goals of the NIH roadmap. Our proposal is consistent with the first goal of Healthy People 2010, to increase quality and years of healthy life, and covers several of its focus areas including Mental Health and Mental Disorders, Disability and Secondary Conditions, Diabetes, Nutrition and Overweight.